A plurality of proposals have been made about a clustering mechanism that allows high-speed printing by distributing output destinations upon printing a large number of pages or copies. A device (printer, multi-functional peripheral equipments, or the like) on the network distributes its processes to other network-connected devices to execute distributed processing, thus rapidly completing the processes.
The distributed processing based on the clustering mechanism divides pages or copies to be printed and makes a plurality of devices execute processes up to printing of the divided pages or copies, and a device designated by a client does not print all pages or copies. Therefore, such processing is suited to parallelly print a large quantity of printed matters. However, upon executing a finishing process such as page sorting or the like, printed matters separately output by a plurality of devices must be collected using arbitrary means, and must then be processed and finished.
As distributed processing associated with printing, Japanese Patent Application Laid-open No. 9-204277 discloses a technique which improves the processing speed of the entire system by efficiently collecting print data rasterized by a plurality of devices on the network. Also, Japanese Patent Application Laid-open No. 9-218861 discloses a technique which shortens an idle time (i.e., the control waits for another job while holding intermediate data of processing) upon scheduling parallel processing of printer description language (PDL) data.
When the distributed load system of grid computing is applied to computer devices such as printers, multi-functional peripheral equipment, and the like connected to a computer network, if a device that inputs a job which is received from a client and includes a large number of pages or requires a long processing time is executing another job (e.g., a print job received from a host computer), the distributed processing requires much time, and the client user cannot obtain the print result soon.